Integrative genomic analysis of methylphenidate response in attention-deficit/hyperactivity disorder

www.nature.com/scientificreports OPEN Received: 31 May 2017 Accepted: 15 January 2018 Published: xx xx xxxx Integrative genomic analysis of methylphenidate response in attention-deficit/hyperactivity disorder Mireia Pagerols1,2, Vanesa Richarte2,3,4, Cristina Sánchez-Mora1,2,3, Paula Rovira1,2, María Soler Artigas1,3, Iris Garcia-Martínez1,2, Eva Calvo-Sánchez1,2, Montse Corrales2,4, Bruna Santos da Silva5, Nina Roth Mota6,7, Marcelo Moraes Victor7, Luis Augusto Rohde7,8, Eugenio Horacio Grevet7,8, Claiton Henrique Dotto Bau5,7, Bru Cormand9,10,11,12, Miguel Casas1,2,3,4, Josep Antoni Ramos-Quiroga1,2,3,4 & Marta Ribasés1,2,3 Methylphenidate (MPH) is the most frequently used pharmacological treatment in children with attention-deficit/hyperactivity disorder (ADHD). However, a considerable interindividual variability exists in clinical outcome. Thus, we performed a genome-wide association study of MPH efficacy in 173 ADHD paediatric patients. Although no variant reached genome-wide significance, the set of genes containing single-nucleotide polymorphisms (SNPs) nominally associated with MPH response (P < 0.05) was significantly enriched for candidates previously studied in ADHD or treatment outcome. We prioritised the nominally significant SNPs by functional annotation and expression quantitative trait loci (eQTL) analysis in human brain, and we identified 33 SNPs tagging cis-eQTL in 32 different loci (referred to as eSNPs and eGenes, respectively). Pathway enrichment analyses revealed an over-representation of genes involved in nervous system development and function among the eGenes. Categories related to neurological diseases, psychological disorders and behaviour were also significantly enriched. We subsequently meta-analysed the association with clinical outcome for the 33 eSNPs across the discovery sample and an independent cohort of 189 ADHD adult patients (target sample) and we detected 15 suggestive signals. Following this comprehensive strategy, our results provide a better understanding of the molecular mechanisms implicated in MPH treatment effects and suggest promising candidates that may encourage future studies. Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental disorder characterised by persistent and age-inappropriate symptoms of inattention, hyperactivity and/or impulsivity1, which significantly impacts on academic, social, emotional and psychological functioning. With a worldwide prevalence ranging from 5.3 to 7.1% in school-age children and adolescents2, ADHD is one of the most common childhood 1 Psychiatric Genetics Unit, Group of Psychiatry, Mental Health and Addiction, Vall d’Hebron Research Institute (VHIR), Universitat Autònoma de Barcelona, Barcelona, Spain. 2Department of Psychiatry, Hospital Universitari Vall d’Hebron, Barcelona, Spain. 3Biomedical Network Research Centre on Mental Health (CIBERSAM), Instituto de Salud Carlos III, Barcelona, Spain. 4Department of Psychiatry and Legal Medicine, Universitat Autònoma de Barcelona, Barcelona, Spain. 5Department of Genetics, Institute of Biosciences, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil. 6Department of Human Genetics and Psychiatry, Donders Institute for Brain, Cognition and Behaviour, Radboud University Medical Centre, Nijmegen, The Netherlands. 7ADHD Outpatient Program, Adult Division, Hospital de Clínicas de Porto Alegre, Porto Alegre, Brazil. 8Department of Psychiatry, Faculty of Medicine, Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil. 9Departament de Genètica, Microbiologia i Estadística, Facultat de Biologia, Universitat de Barcelona, Barcelona, Spain. 10Centro de Investigación Biomédica en Red de Enfermedades Raras (CIBERER), Instituto de Salud Carlos III, Barcelona, Spain. 11Institut de Biomedicina de la Universitat de Barcelona (IBUB), Barcelona, Spain. 12Institut de Recerca Sant Joan de Déu (IR-SJD), Esplugues de Llobregat, Spain. Correspondence and requests for materials should be addressed to M.R. (email: marta.ribases@ vhir.org) SCIENTIFIC REPORTS | (2018) 8:1881 | DOI:10.1038/s41598-018-20194-7 1 www.nature.com/scientificreports/ psychiatric conditions and causes high costs to the healthcare system and society3,4. Although its aetiology is largely unknown, several family, twin and adoption studies reported heritability estimates around 76%5, suggesting a strong genetic component in the pathogenesis of the disorder. Among the wide variety of pharmacological options available in ADHD treatment, methylphenidate (MPH) is the first-line choice in paediatric patients, given its proved general efficacy in reducing ADHD symptoms and improving neuropsychological performance on executive functions6,7. However, a considerable interindividual variability exists in clinical outcome, optimal dosage and duration of effect8,9, which may reflect underlying genetic influences. Most of the pharmacogenetic studies conducted so far in ADHD patients have focused on genes related to the catecholamine neurotransmission, with SLC6A3 and DRD4 being the most extensively investigated, since MPH is thought to exert its therapeutic effects through the inhibition of the dopamine and the norepinephrine transporters10. Based on this putative mechanism of action, additional genes such as DRD2, DRD5, COMT, SLC6A2, ADRA2A, TPH2, SLC6A4, HTR1B, HTR2A and MAOA11 have been considered plausible candidates that may influence medication response. Nevertheless, a recent review on ADHD pharmacogenetics in childhood reported no consistent effects for dopaminergic and serotoninergic signaling, and suggested neurodevelopmental genes as new promising targets12. Given that candidate gene-based investigations have not reached many compelling results, genome-wide association studies (GWAS) may represent an alternative, hypothesis-free approach to unravel the molecular mechanisms implicated in MPH treatment. To date, only one prior GWAS evaluated the efficacy of a MPH transdermal system in 187 children with ADHD13. Although no genome-wide significant associations were found, the metabotropic glutamate receptor 7 (GRM7) and two SNPs within the SLC6A2 gene showed potential involvement in MPH response. Using that sample, Mick et al.14 conducted a secondary GWAS of changes in blood pressure after MPH therapy and detected nominal evidence for genes functionally related to blood pressure regulation and other cardiovascular phenotypes, including a SNP in a K+-dependent Na+/Ca2+ exchanger (SLC24A3). Furthermore, despite the fact that GWAS have been useful to identify genetic risk loci for multiple complex conditions, yet the functional effects of the trait-associated variants and the underlying pathological mechanisms remain mainly elusive. Based on the absence of clear conclusions regarding MPH response raised by previous genetic studies, we undertook a GWAS of MPH efficacy in 173 ADHD paediatric patients and, for the first time to our knowledge, we integrated data from functional annotation, expression quantitative trait loci (eQTL) and enrichment analyses to characterise the biological pathways associated with treatment response. Additionally, we performed a polygenic risk score analysis and a meta-analysis across the study sample and an independent population of 189 ADHD adult patients. Materials and Methods Discovery population. The study sample included 173 ADHD paediatric patients for whom MPH was prescribed. Subjects were required to satisfy full DSM-IV criteria for ADHD, be under 18 years of age, Spanish of Caucasian origin and have never received MPH treatment. Patients with an IQ below 70 or having pervasive developmental disorders were not eligible for the investigation. Additional exclusion criteria included schizophrenia or other psychotic disorders; adoption; sexual or physical abuse; birth weight <1.5 kg; any significant neurological or systemic disease that might explain ADHD symptoms; and clinical contra-indication to MPH. Comorbid oppositional defiant disorder, conduct disorder, depression and anxiety disorders were allowed unless determined to be the primary cause of ADHD symptomatology. The study was approved by the Ethics Committee of the Hospital Universitari Vall d’Hebron and all methods were performed in accordance with the relevant guidelines and regulations. Written informed consent was obtained from parents/caregivers. Clinical assessment. Diagnoses of ADHD and comorbidities were established by child psychiatrists blind to patients’ genotypes through the Present and Lifetime version of the Kiddie Schedule for Affective Disorders and Schizophrenia for School-Age Children (K-SADS-PL). Furthermore, families were interviewed with the Clinical Global Impression-Severity scale (CGI-S). Additional information on clinical assessment is available elsewhere15. Pharmacological intervention. Patients were treated according to the program’s recommendations of initiating treatment with MPH at low to moderate dose and titrating to higher doses until no further clinical improvement or limiting adverse effects were observed. The mean daily dose of MPH prescribed was 1.06 mg/kg (SD = 0.28). Risperidone was the most frequent concomitant drug. Treatment outcome. We considered the Clinical Global Impression-Improvement scale (CGI-I)16, which ranges from 1 (‘very much improved’) to 7 (‘very much worse’), as the primary outcome measure of treatment success. Those patients rated with a CGI-I score of two points or less after eight weeks of treatment were considered as responders, while the remaining were classified as non-responders. Genome-wide association study. Genomic DNA was isolated from peripheral blood leukocytes by a salting out procedure17. A total of 173 samples were genotyped on the Infinium PsychArray-24 BeadChip platform (Illumina, San Diego, CA, USA), which covers 588,628 markers, and processed at the Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard (Cambridge, MA, USA). Pre-imputation quality control and principal components analysis were implemented following the QC and PCA modules from the Ricopili with the default settings (https://sites.google.com/a/broadinstitute.org/ricopili/). Genotype imputation was performed with the pre-phasing and imputation strategy using the EUR population of the 1,000 Genomes Project Phase 1 dataset as the reference panel (http://www.1000genomes.org/). We assured the accuracy of the imputation SCIENTIFIC REPORTS | (2018) 8:1881 | DOI:10.1038/s41598-018-20194-7 2 www.nature.com/scientificreports/ data by filtering best-guess genotypes for an info score <0.3. This resulted in a total of 11,051,824 markers eligible for association tests. Before GWAS analysis, further quality control measures were applied using the PLINK software18. Individuals exhibiting high rates of genotype missingness (>98%) were removed, as well as SNPs with low call rate (<0.99), MAF < 0.01 or failing Hardy-Weinberg equilibrium test (P < 1e-06). Finally, 173 subjects and 3,566,199 variants were tested for association with MPH response through logistic regression under an additive model, which included those clinical variables (i.e., CGI-S baseline scores) and principal components (i.e., PC6) significantly associated with clinical outcome (P ≤ 0.05) as covariates. Identification of candidate causal SNPs. Among the SNPs showing nominal association with treatment outcome (P < 0.05), we used the genome pipeline of SNPinfo (http://snpinfo.niehs.nih.gov/)19 to prioritise those that were more likely to affect protein sequence, transcriptional regulation, mRNA splicing or miRNA binding based on functional annotation. GenomePipe parameter values included: GWAS population = CEU; study population = CEU; flanking region = 200,000 bp; GWAS P-value < 0.05; LD threshold = 0.8; and MAF = 0.01 for all prediction methods. Additionally, we combined both the predicted conserved transcription factor-binding sites (TFBS) with the regulatory potential score (RP Score; available at http://genome.ucsc.edu) to improve predictions as suggested in several studies20–22. Cis-expression quantitative trait loci analysis. Cis-eQTL analysis was conducted on 193 neuropathologically normal cortical samples of adult humans from Myers et al.23. Expression-genotype pairs were extracted after extending the genotyped data by imputation as previously described, and considering a 10 kb window around the untranslated regions. Rank-invariant normalised expression levels were log10 transformed and transcripts detected in less than 75% of the samples were discarded from the study. Association tests were performed under a linear model with the MatrixEQTL R Package24, including gender, age at death, cortical region, day of expression hybridisation, institute source of sample, post-mortem interval and transcripts detected rate in each sample as covariates. Functional and pathway enrichment analysis. The biological functions and pathways related to genes containing at least one SNP nominally associated with both MPH response and human cortical expression levels (referred to as eSNPs) were assessed through the Ingenuity Pathway Analysis software (IPA) (Ingenuity Systems, Redwood City, CA, USA; www.ingenuity.com). IPA was also used to test for over-representation of genes previously studied in either ADHD or treatment outcome. Candidate genes for ADHD or MPH response were selected based on the gene list provided by the ADHDgene database (http://adhd.psych.ac.cn/index.do)25 and a comprehensive search for published reviews of ADHD genetic and pharmacogenetic studies11,12,26–31. Thus, a total of 436 genes were considered (Supplementary Table S1). Fisher’s exact tests, with a Benjamini-Hochberg-adjusted P-value (PB-H) < 0.05 as significance threshold, were applied in all analyses. To achieve meaningful statistics and interpretation of the results, we restricted the enrichment analysis to those annotation terms that included ≥2 genes of our dataset. Polygenic risk score analysis. We generated polygenic risk scores (PRS) based on the results of the present GWAS using the Polygenic Risk Score software (PRSice)32. A logistic regression model was applied to test whether PRS at multiple stepwise P-value thresholds (i.e., PT < 1e-04, PT < 1e-03, PT < 0.05, PT < 0.1, PT < 0.2, PT < 0.3, PT < 0.4, and PT < 0.5) predicted treatment outcome in an independent sample of patients with ADHD (target population). The target population comprised 189 Brazilian adults from the Adult ADHD Outpatient Clinic of the Hospital de Clínicas de Porto Alegre, who underwent immediate-release MPH treatment. Diagnoses of ADHD and comorbidities, as well as inclusion/exclusion criteria, were achieved as previously described33. The outcome measures of MPH treatment were the CGI-S scale, applied before medication and at least four weeks after its beginning, and the CGI-I scale. Drug response was defined following the criteria used in the discovery sample. Similarly, samples were genotyped and imputed using the same platform, imputation protocol and reference panel. The resulting dataset consisted of 7,304,149 SNPs with an info score >0.6, a genotype call probability >0.8 and a missing rate <0.02. Potential confounders were included as covariates in the PRS model if they were associated with MPH response (P ≤ 0.05) in the target sample (i.e., CGI-S baseline scores, use of concomitant medication and presence of phobia as comorbid condition), as well as the 10 first principal components to control for population stratification. Meta-analysis. The eSNPs nominally associated with MPH response in the discovery sample were meta-analysed across the discovery and the target population used in the PRS analysis by the inverse-variance weighted method. The threshold for significance was set at P ≤ 1.52e-03 under the more conservative Bonferroni correction, taking into account 33 SNPs. Data availability. The datasets generated and/or analysed during the current study are not publicly available due to ethics constraints but are available from the corresponding author on reasonable request. Results Genome-wide association study in the discovery population. Subjects were predominantly male (84.4%), with an average age at assessment of 9.59 (SD = 2.91) years (range 5–17). One hundred and thirty-one participants (75.7%) met DSM-IV criteria for ADHD-combined subtype, 37 (21.4%) had ADHD-inattentive subtype and 5 (2.9%) were diagnosed with ADHD-hyperactive-impulsive subtype. Comorbid disorders were present in a modest number of patients (22.5%), the main ones being disabilities in reading and writing (12.7%), SCIENTIFIC REPORTS | (2018) 8:1881 | DOI:10.1038/s41598-018-20194-7 3 www.nature.com/scientificreports/ oppositional defiant disorder (5.8%) and tic disorders (1.7%). One hundred and forty-one subjects (81.5%) responded favourably to treatment according to the CGI-I scale, while 32 (18.5%) failed to show a clinical response to MPH. Responders and non-responders were comparable with regard to age, sex, ADHD subtype, comorbidity, use of concomitant medication, MPH dose and drug formulation (P > 0.05). There were significant differences, however, in the severity of symptoms as assessed by the CGI-S scale (P < 1e-03), with children resistant to MPH scoring higher at the baseline evaluation than children showing clinical improvement (Supplementary Table S2). No variant reached genome-wide significance (P < 5e-08). However, the set of 4,709 genes containing SNPs nominally associated with MPH response (P < 0.05; Supplementary Table S3) was significantly enriched for candidates previously studied in ADHD or treatment outcome, with 199 out of 436 being present in this category (ratio = 0.46; PB-H = 1.56e-31). Identification of candidate causal SNPs and cis-expression quantitative trait loci analysis. Considering these results, we prioritised the SNPs with P-values below 0.05 based on functional annotation and eQTL analysis rather than focusing on the top significant hits. Eight hundred and ninety-six independent markers were selected as candidate causal variants by functional annotation (Supplementary Table S4) and were subjected to further cis-eQTL analysis on a pre-existing dataset of 193 neuropathologically normal human cortical samples23. After imputation and quality control, a total of 284 variants and 300 genes with detectable expression levels in at least 75% of the samples were available for 146 individuals. Of these, we identified 33 SNPs tagging cis-eQTL in 32 different loci (referred to as eGenes), with eight SNP-gene pairs surpassing the 0.05 false discovery rate (FDR) threshold: rs12302749-SPSB2, PFDR = 1.13e-05; rs1061115-PYROXD2, PFDR = 2.17e-04; rs2071421-ARSA, PFDR = 7.26e-04; rs11553441-RRP7A, PFDR = 7.26e-04; rs4902333-CHURC1, PFDR = 7.26e-04; rs17279558-GGH, PFDR = 0.013; rs9901673-SENP3, PFDR = 0.023; and rs17685420-PEBP4, PFDR = 0.041 (Table 1). Functional and pathway enrichment analysis. The set of 32 eGenes included three candidates previously investigated in ADHD, namely ALDH1L134, CDH2335 and CMTM836 (ratio = 0.007; PB-H = 0.023), and showed over-representation of genes implicated in abnormal morphology of molecular layer of cerebellum (PB-H = 0.012), abnormal morphology of white matter (PB-H = 0.012), morphology of axons (PB-H = 0.012), morphology and length of neurites (PB-H = 0.012 and PB-H = 0.021, respectively), coordination (PB-H = 0.022), and formation of hippocampus (PB-H = 0.033). Interestingly, categories related to neurological diseases, psychological disorders and behaviour were also significantly enriched, including learning deficit (PB-H = 0.012), hyperactive behaviour (PB-H = 0.015) and spatial learning (PB-H = 0.018) (Table 2). Polygenic risk score analysis and meta-analysis using the target population. Finally, in order to assess the predictive value of our findings we first computed PRS derived from the present GWAS in an independent sample of ADHD adult patients for whom data on response to MPH were available. The demographic and clinical characteristics of the target population according to the response status are presented in Supplementary Table S5. Briefly, 85.2% of subjects (n = 161) were classified as responders, while 14.8% (n = 28) exhibited a reduced or lack of improvement. Responders and non-responders significantly differed with regard to CGI-S baseline scores, use of concomitant medication and presence of phobia as comorbid condition, and thus these additional risk factors were entered as covariates in the PRS model, as well as the 10 first principal components to control for population stratification. Since we did not detect significant results at any of the predefined P-value thresholds, we subsequently focused on the 33 eSNPs nominally associated with treatment outcome in the discovery sample and we increased statistical power by performing a meta-analysis across the discovery and the target population. Sixteen suggestive signals were identified (Table 3). Among them, 15 revealed the same direction of effect, with rs17685420 in the PEBP4 gene being significant after Bonferroni correction (OR = 3.07 (1.76–5.35), P = 7.90e-05), followed by additional compelling markers such as rs2071421 within ARSA (OR = 2.63 (1.29– 5.37), P = 7.71e-03), rs2886059 in ALDH1L1 (OR = 2.30 (1.14–4.66), P = 0.020), and rs17712523 in CDH23 (OR = 2.13 (1.07–4.24), P = 0.031). Discussion To our knowledge, this is the first study investigating the genetic basis of MPH response from an integrative perspective that combines GWAS data, functional annotation, eQTL in relevant tissues to ADHD and pathway enrichment analyses. Our results highlight genes related to nervous system development and function, neurological diseases and psychological disorders. Thus, this comprehensive strategy provides a better understanding of the molecular mechanisms implicated in MPH treatment effects and suggests promising candidates that may contribute to clinical outcome. In our attempt to improve earlier genetic studies by bridging the gap between genotype and phenotype, we prioritised the nominally significant SNPs based on functional annotation and cis-eQTL analysis in human brain, and we identified three candidates previously investigated in ADHD: ALDH1L134, CDH2335 and CMTM836. Of these, CMTM8 showed overlapping association between adult ADHD and bipolar disorder36, and ALDH1L1, which yielded suggestive results in the present meta-analysis of MPH response, has been related to other neuropsychiatric conditions such as major depressive disorder or schizophrenia37,38. In addition, the ALDH1L1 locus was found among the top hits of a GWAS conducted on children and adolescents with ADHD34 and contains non-synonymous rare variants identified through whole-exome sequencing in an ADHD nuclear family39. Similarly, CDH23 harbours one of the top SNPs from a pooling-based GWA of adult ADHD35 and reached nominal significance in our meta-analysis. CDH23 is a member of the cadherin superfamily that mediates calcium-dependent cell-cell adhesion. The activity of cadherins depends on their anchorage to the neuronal cytoskeleton through proteins termed catenins (e.g., CTNNA2), which in turn activate KALRN, a key regulator SCIENTIFIC REPORTS | (2018) 8:1881 | DOI:10.1038/s41598-018-20194-7 4 www.nature.com/scientificreports/ eQTL P-value eQTL adjusted P-value (FDR)e 0.115 3.34e-08 1.13e-05 −0.088 1.28e-06 2.17e-04 Gene Chra Start baseb Stop basec SNP SNP based Risk allele OR (95% CI) GWAS P-value SPSB2 12 6870935 6873357 rs12302749 6867132 T 2.31 (1.22–4.39) 0.011 PYROXD2 10 98383565 98415221 rs1061115 98417292 G 2.23 (1.13–4.41) 0.021 ARSA 22 50622754 50628173 rs2071421 50625988 T 2.56 (1.06–6.22) 0.037 0.104 8.26e-06 7.26e-04 RRP7A 22 42508335 42519823 rs11553441 42516091 C 3.13 (1.15–8.54) 0.026 0.177 1.04e-05 7.26e-04 Beta CHURC1 14 64914361 64935368 rs4902333 64909368 T 2.37 (1.24–4.51) 8.73e-03 0.116 1.07e-05 7.26e-04 GGH 8 63015079 63039051 rs17279558 63015187 C 3.61 (1.12–11.7) 0.032 0.130 2.38e-04 0.013 SENP3 17 7561992 7571969 rs9901673 7580783 A 3.95 (1.79–8.71) 6.53e-04 0.052 4.66e-04 0.023 PEBP4 8 22713251 22941095 rs17685420 22927888 T 2.87 (1.38–5.94) 4.62e-03 −0.073 9.72e-04 0.041 STRBP 9 123109494 123268576 rs9032 123104493 C 2.28 (1.07–4.85) 0.033 0.071 2.15e-03 0.081 ETFDH 4 158672101 158708713 rs11559290 158680524 C 2.08 (1.01–4.28) 0.048 −0.048 2.57e-03 0.087 CORO7 16 4354542 4416961 rs3810818 4382028 A 2.10 (1.13–3.94) 0.020 −0.053 3.17e-03 0.098 FXR2 17 7591230 7614897 rs9901675 7581494 A 4.12 (1.32–12.9) 0.015 0.130 3.84e-03 0.107 NFIB 9 14081843 14398983 rs7858 14087770 C 2.89 (1.02–8.19) 0.045 −0.055 4.12e-03 0.107 ALDH1L1 3 126103561 126181526 rs2886059 126146923 C 2.73 (1.04–7.14) 0.041 −0.078 5.31e-03 0.129 OPCML 11 132403361 133532983 rs751655 132623600 C 3.08 (1.18–8.02) 0.022 −0.063 7.43e-03 0.158 PURA 5 140114123 140119416 rs2013169 140118020 T 3.32 (1.23–9.01) 0.018 0.071 7.45e-03 0.158 ZDHHC7 16 84974460 85011732 rs3210967 84975857 C 2.09 (1.11–3.94) 0.023 0.056 8.03e-03 0.160 WRB 21 39380287 39397889 rs3761372 39371919 T 3.38 (1.39–8.22) 7.36e-03 0.071 8.54e-03 0.161 FARP2 2 241356249 241494842 rs757978 241431686 C 5.18 (1.19–22.6) 0.029 0.062 0.010 0.181 SENP3 17 7561992 7571969 rs11552708 7559238 A 3.81 (1.50–9.72) 5.05e-03 0.041 0.011 0.189 ZNF565 19 36182060 36215084 rs4805162 36183403 G 2.24 (1.19–4.22) 0.012 0.034 0.016 0.255 ESYT2 7 158730998 158829628 rs1061735 158733764 G 2.91 (1.10–7.67) 0.031 0.030 0.017 0.255 HTT 4 3074510 3243960 rs362272 3233253 G 2.40 (1.06–5.42) 0.035 −0.033 0.019 0.276 CMTM8 3 32238679 32370325 rs4627790 32259860 C 1.98 (1.07–3.68) 0.030 0.054 0.021 0.298 ZNF134 19 57614219 57624717 rs10413455 57620255 A 5.75 (1.35–24.4) 0.018 0.061 0.024 0.323 PDIA2 16 283118 287209 rs1048786 286916 C 3.55 (1.19–10.6) 0.023 −0.087 0.027 0.345 PIGM 1 160027672 160031993 rs12409352 160030645 A 2.67 (1.00–7.12) 0.049 0.029 0.032 0.395 TRIB3 20 380629 397559 rs2295490 388261 G 2.06 (1.06–4.00) 0.033 0.092 0.034 0.406 ZNF211 19 57633167 57644046 rs10420097 57633193 G 7.29 (1.82–29.3) 5.18e-03 0.084 0.038 0.439 ARHGAP12 10 31805398 31928876 rs2799018 31913141 T 1.89 (1.00–3.56) 0.049 −0.036 0.039 0.446 ARHGEF28 5 73626158 73941993 rs929740 73621913 G 2.52 (1.31–4.84) 5.40e-03 −0.037 0.042 0.453 CDH23 10 71396934 71815947 rs17712523 71777857 G 2.61 (1.05–6.48) 0.039 −0.073 0.045 0.475 ELP5 17 7252053 7259940 rs4562 7260420 A 2.22 (1.24–3.95) 6.95e-03 −0.031 0.048 0.497 Table 1. Cis-associated gene-SNP pairs with a nominal significant effect on methylphenidate response in the GWAS analysis. Note: SNP, single-nucleotide polymorphism; GWAS, genome-wide association study; Chr, gene chromosomal location; OR, odds ratio; CI, confidence interval; eQTL, expression quantitative trait loci. a,b,c,d All relative to the human reference genome GRCh38 (NCBI Build 38). eSignificance threshold for the False Discovery Rate (FDR) correction at P < 0.05. of dendritic spine development and synaptic plasticity underlying learning and memory40. This is of particular interest since catenin-cadherin cell-adhesion complexes are important in cerebellar and hippocampal lamination41 and both CTNNA2 and KALRN have shown nominal associations with clinical outcome in our GWAS. In this sense, Park et al.41 demonstrated that mice lacking the actin-binding domain of Ctnna2 (cdf-mutant mice) exhibited abnormal morphology of cerebellum and hippocampus. Moreover, the cdf-mutant mice showed an impaired control of the startle response and deficits in startle modulation have also been found in ADHD patients42,43. Therefore, cell-adhesion molecules and regulators of synaptic plasticity may play a role in MPH treatment effects, which is in agreement with data from genome-wide linkage and association studies pointing to cadherin13 (CDH13) as one of the most consistent candidates implicated in ADHD pathophysiology. Specifically, CDH13 was detected in three independent GWAS34,35,44 and lies within the 16q22-16q24 region identified in a meta-analysis of seven ADHD linkage scans45. Furthermore, SNPs in this gene have been linked to defects in verbal working memory and hyperactive/impulsive symptoms in subjects with ADHD46,47, addiction vulnerability and drug dependence (e.g., methamphetamine, alcohol, and nicotine)48,49. Pathway enrichment analysis provided further evidence for neuroplastic changes in response to MPH treatment, considering the over-representation of genes involved in morphology of neurons, neuroglia, white matter and brain regions relevant to ADHD (e.g., cerebellum, cerebral cortex, and hippocampus) that we found among eGenes associated with drug response. Our results are in accordance to a wealth of data from neuroimaging studies showing that unmedicated ADHD patients present cortical thickness and reduced white matter volume in several areas of the brain compared to healthy subjects, while medicated children do not differ from control group50–53. In addition to structural alterations, ADHD patients exhibit deficits in neural functioning, which may SCIENTIFIC REPORTS | (2018) 8:1881 | DOI:10.1038/s41598-018-20194-7 5 www.nature.com/scientificreports/ Categories Diseases or functions annotation Adjusted P-value (Benjamini-Hochberg)a Nervous System Development and Function, Organ Morphology, Organismal Development abnormal morphology of 0.012 molecular layer of cerebellum ARSA, PURA Nervous System Development and Function, Organ Morphology, Tissue Morphology abnormal morphology of white matter 0.012 ARSA, PURA Cellular Development, Embryonic Development, Organismal Development differentiation of neuronal progenitor cells 0.012 FXR2, HTT Developmental Disorder, Neurological Disease learning deficit 0.012 ARSA, HTT Cell Morphology, Nervous System Development and Function, Organ Morphology, Organismal Development, Tissue Morphology morphology of granule cells 0.012 HTT, NFIB Cell Morphology, Haematological System Development and Function, Nervous System Development and Function morphology of microglia 0.012 ARSA, HTT Neurological Disease gait disturbance 0.012 ARSA, HTT, PURA Cell Morphology, Nervous System Development and Function, Tissue Morphology morphology of axons 0.012 ARSA, HTT, PURA Cell Morphology, Nervous System Development and Function morphology of neuroglia 0.012 ARSA, HTT, NFIB Cell Morphology, Nervous System Development and Function, Organ Morphology, Organismal Development morphology of brain cells 0.012 ARSA, HTT, NFIB, PURA Cell Morphology, Nervous System Development and Function, Tissue Morphology morphology of neurites 0.012 ARSA, FARP2, HTT, PURA Cell Morphology, Nervous System Development and Function, Tissue Morphology morphology of neurons 0.012 ARSA, CDH23, FARP2, HTT, NFIB, PURA Molecules Neurological Disease late-onset encephalopathy 0.014 ARSA, HTT Psychological Disorders hyperactive behaviour 0.015 ARSA, FXR2, HTT Neurological Disease tremor 0.015 ARSA, HTT, PURA Nervous System Development and Function, Organ Morphology, Organismal Development abnormal morphology of dentate gyrus 0.015 NFIB, PURA Cell Morphology, Nervous System Development and Function, Organ Morphology, Organismal Development, Tissue Morphology abnormal morphology of Purkinje cells 0.015 ARSA, PURA Cell Death and Survival, Cellular Compromise, Neurological Disease, Organismal Injury and Abnormalities, Tissue Morphology neurodegeneration of Purkinje cells 0.015 ARSA, HTT Nervous System Development and Function, Organ Morphology, Organismal Development abnormal morphology of telencephalon 0.015 ARSA, HTT, NFIB Behaviour spatial learning 0.018 ARSA, FXR2, HTT Nervous System Development and Function, Organ Morphology, Organismal Development mass of brain 0.019 HTT, PURA Cell Morphology, Cellular Function and Maintenance, Nervous System Development and Function, Tissue Morphology length of neurites 0.021 FARP2, HTT HTT, NFIB Organismal Injury and Abnormalities abnormality of head 0.022 Nervous System Development and Function coordination 0.022 ARSA, FXR2, HTT Cellular Development differentiation of stem cells 0.022 FXR2, HTT, NFIB Developmental Disorder, Neurological Disease, Organismal Injury and Abnormalities cerebral dysgenesis 0.022 NFIB, PURA Nervous System Development and Function, Organ Morphology, Organismal Development morphology of cerebral cortex 0.023 HTT, NFIB, PURA Organismal Development size of head 0.024 HTT, NFIB, PURA Neurological Disease, Organismal Injury and Abnormalities astrocytosis 0.025 ARSA, HTT Cell Death and Survival, Cellular Compromise, Neurological Disease, Tissue Morphology neurodegeneration of axons 0.026 ARSA, HTT Cellular Growth and Proliferation, Nervous System Development and Function, Organ Development proliferation of brain cells 0.030 HTT, PURA Nervous System Development and Function, Organ Morphology, Organismal Development abnormal morphology of brain 0.030 ARSA, HTT, NFIB, PURA Embryonic Development, Organismal Development, Tissue Development mesoderm development 0.032 CHURC1, HTT Embryonic Development, Nervous System Development and Function, Organ Development, Organismal Development, Tissue Development formation of hippocampus 0.033 HTT, NFIB Nervous System Development and Function, Organ Morphology, Tissue Morphology quantity of brain cells 0.042 HTT, PURA Nervous System Development and Function sensation 0.047 CDH23, FXR2, HTT Table 2. Significantly enriched biological functions and diseases identified by Ingenuity Pathway Analysis within the eGenes associated with methylphenidate response. Note: eGenes, genes whose expression levels are associated with at least one genetic variant. aSignificance threshold for the Benjamini-Hochberg correction at P < 0.05. SCIENTIFIC REPORTS | (2018) 8:1881 | DOI:10.1038/s41598-018-20194-7 6 www.nature.com/scientificreports/ SPAIN BRAZIL META-ANALYSIS SNP Chra SNP baseb Risk allele OR (95% CI) P-value OR (95% CI) P-value OR (95% CI) P-valuec Gene rs17685420 8 22927888 T 2.87 (1.38–5.94) 4.62e-03 3.38 (1.43–8.01) 5.71e-03 3.07 (1.76–5.35) 7.90e-05 PEBP4 rs3210967 16 84975857 C 2.09 (1.11–3.94) 0.023 2.25 (1.02–4.94) 0.044 2.15 (1.31–3.52) 2.40e-03 ZDHHC7 rs10413455 19 57620255 A 5.75 (1.35–24.4) 0.018 3.86 (0.63–23.7) 0.144 4.93 (1.59–15.3) 5.70e-03 ZNF134 rs2071421 22 50625988 T 2.56 (1.06–6.22) 0.037 2.76 (0.84–9.15) 0.096 2.63 (1.29–5.37) 7.71e-03 ARSA rs12302749 12 6867132 T 2.31 (1.22–4.39) 0.011 1.49 (0.72–3.10) 0.280 1.91 (1.18–3.09) 8.48e-03 SPSB2 rs10420097 19 57633193 G 7.29 (1.82–29.3) 5.18e-03 1.58 (0.14–17.7) 0.712 4.98 (1.49–16.6) 9.13e-03 ZNF211 rs3810818 16 4382028 A 2.10 (1.13–3.94) 0.020 1.40 (0.62–3.16) 0.413 1.81 (1.10–2.97) 0.019 CORO7 rs2886059 3 126146923 C 2.73 (1.04–7.14) 0.041 1.89 (0.67–5.33) 0.230 2.30 (1.14–4.66) 0.020 ALDH1L1 rs9901675 17 7581494 A 4.12 (1.32–12.9) 0.015 1.57 (0.33–7.46) 0.572 2.95 (1.18–7.39) 0.021 FXR2 rs4805162 19 36183403 G 2.24 (1.19–4.22) 0.012 1.21 (0.59–2.50) 0.608 1.72 (1.07–2.76) 0.026 ZNF565 rs4562 17 7260420 A 2.22 (1.24–3.95) 6.95e-03 1.05 (0.51–2.15) 0.889 1.65 (1.05–2.59) 0.029 ELP5 rs17712523 10 71777857 G 2.61 (1.05–6.48) 0.039 1.63 (0.57–4.66) 0.366 2.13 (1.07–4.24) 0.031 CDH23 rs2799018 10 31913141 T 1.89 (1.00–3.56) 0.049 1.40 (0.66–2.97) 0.375 1.67 (1.03–2.71) 0.038 ARHGAP12 rs12409352 1 160030645 A 2.67 (1.00–7.12) 0.049 1.63 (0.54–4.95) 0.387 2.15 (1.03–4.49) 0.041 PIGM rs4902333 14 64909368 T 2.37 (1.24–4.51) 8.73e-03 0.94 (0.41–2.18) 0.893 1.68 (1.01–2.80) 0.046 CHURC1 rs2295490 20 388261 G 2.06 (1.06–4.00) 0.033 1.22 (0.48–3.13) 0.678 1.73 (1.01–2.98) 0.048 TRIB3 rs4627790 3 32259860 C 1.98 (1.07–3.68) 0.030 1.13 (0.52–2.45) 0.761 1.59 (0.98–2.59) 0.060 CMTM8 rs1048786 16 286916 C 3.55 (1.19–10.6) 0.023 1.15 (0.38–3.48) 0.805 2.03 (0.93–4.43) 0.074 PDIA2 rs9901673 17 7580783 A 3.95 (1.79–8.71) 6.53e-04 0.21 (0.054–0.77) 0.019 1.82 (0.92–3.59) 0.084 SENP3 rs751655 11 132623600 C 3.08 (1.18–8.02) 0.022 0.99 (0.37–2.62) 0.986 1.76 (0.89–3.49) 0.104 OPCML rs11559290 4 158680524 C 2.08 (1.01–4.28) 0.048 1.03 (0.39–2.74) 0.957 1.62 (0.90–2.90) 0.105 ETFDH rs7858 9 14087770 C 2.89 (1.02–8.19) 0.045 1.12 (0.41–3.09) 0.820 1.78 (0.86–3.67) 0.119 NFIB rs929740 5 73621913 G 2.52 (1.31–4.84) 5.40e-03 0.69 (0.34–1.41) 0.313 1.40 (0.87–2.27) 0.169 ARHGEF28 rs9032 9 123104493 C 2.28 (1.07–4.85) 0.033 0.71 (0.25–2.01) 0.515 1.52 (0.82–2.81) 0.179 STRBP rs11552708 17 7559238 A 3.81 (1.50–9.72) 5.05e-03 0.11 (0.020–0.62) 0.012 1.70 (0.75–3.86) 0.207 SENP3 rs1061735 7 158733764 G 2.91 (1.10–7.67) 0.031 0.79 (0.32–1.98) 0.618 1.46 (0.75–2.84) 0.265 ESYT2 rs11553441 22 42516091 C 3.13 (1.15–8.54) 0.026 0.74 (0.29–1.90) 0.529 1.46 (0.73–2.90) 0.284 RRP7A rs362272 4 3233253 G 2.40 (1.06–5.42) 0.035 0.72 (0.31–1.68) 0.451 1.34 (0.75–2.41) 0.324 HTT rs757978 2 241431686 C 5.18 (1.19–22.6) 0.029 0.74 (0.22–2.42) 0.613 1.59 (0.63–4.01) 0.325 FARP2 rs17279558 8 63015187 C 3.61 (1.12–11.7) 0.032 0.37 (0.081–1.74) 0.210 1.56 (0.62–3.97) 0.348 GGH rs3761372 21 39371919 T 3.38 (1.39–8.22) 7.36e-03 0.65 (0.30–1.39) 0.268 1.31 (0.73–2.33) 0.365 WRB rs1061115 10 98417292 G 2.23 (1.13–4.41) 0.021 0.58 (0.28–1.24) 0.161 1.22 (0.74–2.02) 0.438 PYROXD2 rs2013169 5 140118020 T 3.32 (1.23–9.01) 0.018 0.28 (0.11–0.73) 9.33e-03 0.92 (0.46–1.84) 0.813 PURA Table 3. Meta-analysis of the eSNPs nominally associated with methylphenidate response across the discovery and the target population. Note: eSNP, single-nucleotide polymorphism associated with cortical expression levels; Chr, gene chromosomal location; OR, odds ratio; CI, confidence interval. a,bAll relative to the human reference genome GRCh38 (NCBI Build 38). cSignificance threshold for Bonferroni correction at P ≤ 1.52e-03. be normalised by MPH. In this sense, Rubia et al.54–56 demonstrated that MPH restores the aberrant activation and functional connectivity in attention, motivation and interference inhibition networks, as well as during error processing, thus improving neuropsychological performance of subjects with ADHD. It should also be noted that 15 out of the 32 eGenes included in the pathway enrichment analysis harboured variants which provided preliminary evidence for an association with clinical outcome across the discovery and an independent sample. Our top hit from the meta-analysis, rs17685420, is located in the phosphatidylethanolamine binding protein 4 (PEBP4), a member of an evolutionary conserved family of proteins with pivotal biological functions such as cell proliferation and survival, stimulation of acetylcholine synthesis and inhibition of serine proteases57. Given that serine proteases are implicated in many processes during development and tissue homeostasis (e.g., neuronal outgrowth, cell migration, and cell death), disturbances in their activity on the nervous system have been proposed as a possible pathological mechanism for neurological disorders58. Indeed, Hohman et al.59 identified a gene-gene interaction involving PEBP4 associated with late onset Alzheimer’s disease (AD) across 13 independent datasets, and decreased expression levels have been found in hippocampus of both AD patients and mouse models for another phosphatidylethanolamine binding protein, the PEBP160–62, which has also shown alterations after methamphetamine and morphine administration63,64. Additional compelling results were detected for ARSA, SPSB2, CORO7 and PIGM. The ARSA gene encodes the arylsulfatase A, whose deficiency is characterised by decline in school performance, emergence of behavioural problems and neurologic symptoms, such as cerebellar ataxia, among others65. SPSB2 has been associated with borderline personality disorder in a genome-wide methylation analysis66 and CORO7, which has shown to be important in brain development67, was identified as a novel candidate gene for emotionality by comparing the expression profile of two mouse lines SCIENTIFIC REPORTS | (2018) 8:1881 | DOI:10.1038/s41598-018-20194-7 7 www.nature.com/scientificreports/ with either high or low anxiety-related behaviour68. Finally, mutations in the PIGM gene, which encodes a protein involved in the synthesis of the glycosylphosphatidylinositol anchor, have been reported in individuals with severe neurological features, including seizures, muscular hypotonia and intellectual disability69. Another interesting finding arising from our research is the significant enrichment for candidates previously related to ADHD or MPH response detected among the set of genes nominally associated with treatment outcome. It is worth mentioning that four of these candidates, namely CTNNA2 (rs79067553, P = 3.51e-05), PARD3B (rs62172701, P = 3.28e-04), LRP1B (rs410870, P = 4.00e-04) and GRM7 (rs17047590, P = 6.36e-04), were significant at P < 1e-03 in the present GWAS analysis. In particular, the metabotropic glutamate receptor 7 (GRM7), which is widely expressed in brain regions relevant to ADHD such as the cerebral cortex, the hippocampus and the cerebellum51,70 and has been associated with the disorder71–73, was also found among the top hits in a prior GWAS of MPH efficacy13, thus supporting the role of the glutaminergic system as a moderator of treatment outcome. The main strengths of our design include the coverage of a considerably higher number of genetic variants in comparison with the study from Mick et al.13 (319,722 vs 3,566,199 markers), the use of an integrative approach that combines GWAS data with bioinformatic methods, and the follow up of our top signals in an independent cohort, which did increase the association of a number of markers located in loci with biologically plausible functions (PEBP4, ARSA, and SPSB2). Nevertheless, some limitations should also be considered when interpreting these results. Given the limited sample size, the present study might not be sufficiently powered to detect individual variants of modest effects and we did not identify any loci reaching the genome-wide threshold. This constraint, however, is heavily conditioned on the difficulty to find the required phenotype as shown by the sample size of the studies included in the last meta-analysis of candidate gene-based investigations on MPH response74. The small dimension of our paediatric sample could also explain the lack of significance of the PRS derived from the GWAS results in an independent population of ADHD adult patients. Alternatively, this discrepancy may be attributed to differences in the genetic background and the clinical heterogeneity (e.g., comorbidities, frequency of clinical subtypes, and sex ratio) of ADHD among children and adults, as suggested by most of the pharmacogenetic studies conducted in adult samples, which failed to replicate variants previously identified in children and adolescents75. Additional methodological aspects or distinct environmental influences between the discovery and the target population may also be responsible for the absence of association. In conclusion, despite not reaching any genome-wide significant association, our results are consistent with previous findings and highlight genes related to morphological abnormalities in brain regions important for motor control, attention and memory, thus supporting the use of bioinformatic and biological evidence as a complement to GWAS data to disentangle the genetic basis of MPH response. References 1. American Psychiatric Association. Diagnostic And Statistical Manual Of Mental Disorders, 5th Edition. (American Psychiatric Publishing Association, 2013). 2. Polanczyk, G. V., Willcutt, E. G., Salum, G. A., Kieling, C. & Rohde, L. A. ADHD prevalence estimates across three decades: an updated systematic review and meta-regression analysis. Int J Epidemiol. 43, 434–442 (2014). 3. Doshi, J. A. et al. Economic impact of childhood and adult attention-deficit/hyperactivity disorder in the United States. J Am Acad Child Adolesc Psychiatry. 51, 990–1002 e2 (2012). 4. Le, H. H. et al. Economic impact of childhood/adolescent ADHD in a European setting: the Netherlands as a reference case. Eur Child Adolesc Psychiatry. 23, 587–598 (2014). 5. Faraone, S. V. & Mick, E. Molecular genetics of attention deficit hyperactivity disorder. Psychiatr Clin North Am. 33, 159–180 (2010). 6. Blum, N. J., Jawad, A. F., Clarke, A. T. & Power, T. J. Effect of osmotic-release oral system methylphenidate on different domains of attention and executive functioning in children with attention-deficit-hyperactivity disorder. Dev Med Child Neurol. 53, 843–849 (2011). 7. Greenhill, L. et al. Guidelines and algorithms for the use of methylphenidate in children with Attention-Deficit/ Hyperactivity Disorder. J Atten Disord. 6(Suppl 1), S89–100 (2002). 8. Charach, A., Ickowicz, A. & Schachar, R. Stimulant treatment over five years: adherence, effectiveness, and adverse effects. J Am Acad Child Adolesc Psychiatry. 43, 559–567 (2004). 9. Wolraich, M. L. & Doffing, M. A. Pharmacokinetic considerations in the treatment of attention-deficit hyperactivity disorder with methylphenidate. CNS Drugs. 18, 243–250 (2004). 10. Wilens, T. E. Effects of methylphenidate on the catecholaminergic system in attention-deficit/hyperactivity disorder. J Clin Psychopharmacol. 28, S46–53 (2008). 11. Kieling, C., Genro, J. P., Hutz, M. H. & Rohde, L. A. A current update on ADHD pharmacogenomics. Pharmacogenomics. 11, 407–419 (2010). 12. Bruxel, E. M. et al. ADHD pharmacogenetics across the life cycle: New findings and perspectives. Am J Med Genet B Neuropsychiatr Genet. 165B, 263–282 (2014). 13. Mick, E., Neale, B., Middleton, F. A., McGough, J. J. & Faraone, S. V. Genome-wide association study of response to methylphenidate in 187 children with attention-deficit/hyperactivity disorder. Am J Med Genet B Neuropsychiatr Genet. 147B, 1412–1418 (2008). 14. Mick, E., McGough, J. J., Middleton, F. A., Neale, B. & Faraone, S. V. Genome-wide association study of blood pressure response to methylphenidate treatment of attention-deficit/hyperactivity disorder. Prog Neuropsychopharmacol Biol Psychiatry. 35, 466–472 (2011). 15. Pagerols, M. et al. Pharmacogenetics of methylphenidate response and tolerability in attention-deficit/hyperactivity disorder. Pharmacogenomics J. 17, 98–104 (2017). 16. Guy, W. ECDEU Assessment Manual For Psychopharmacology, Revised. (US Department of Health, Education and Welfare, 1976). 17. Miller, S. A., Dykes, D. D. & Polesky, H. F. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res. 16, 1215 (1988). 18. Purcell, S. et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 81, 559–575 (2007). 19. Xu, Z. & Taylor, J. A. SNPinfo: integrating GWAS and candidate gene information into functional SNP selection for genetic association studies. Nucleic Acids Res. 37, W600–605 (2009). 20. Elnitski, L. et al. Distinguishing regulatory DNA from neutral sites. Genome Res. 13, 64–72 (2003). SCIENTIFIC REPORTS | (2018) 8:1881 | DOI:10.1038/s41598-018-20194-7 8 www.nature.com/scientificreports/ 21. Elnitski, L., Jin, V. X., Farnham, P. J. & Jones, S. J. Locating mammalian transcription factor binding sites: a survey of computational and experimental techniques. Genome Res. 16, 1455–1464 (2006). 22. King, D. C. et al. Evaluation of regulatory potential and conservation scores for detecting cis-regulatory modules in aligned mammalian genome sequences. Genome Res. 15, 1051–1060 (2005). 23. Myers, A. J. et al. A survey of genetic human cortical gene expression. Nat Genet. 39, 1494–1499 (2007). 24. Shabalin, A. A. Matrix eQTL: ultra fast eQTL analysis via large matrix operations. Bioinformatics. 28, 1353–1358 (2012). 25. Zhang, L. et al. ADHDgene: a genetic database for attention deficit hyperactivity disorder. Nucleic Acids Res. 40, D1003–1009 (2012). 26. Asherson, P. & Gurling, H. Quantitative and molecular genetics of ADHD. Curr Top Behav Neurosci. 9, 239–272 (2012). 27. Bonvicini, C., Faraone, S. V. & Scassellati, C. Attention-deficit hyperactivity disorder in adults: A systematic review and metaanalysis of genetic, pharmacogenetic and biochemical studies. Mol Psychiatry. 21, 872–884 (2016). 28. Froehlich, T. E., McGough, J. J. & Stein, M. A. Progress and promise of attention-deficit hyperactivity disorder pharmacogenetics. CNS Drugs. 24, 99–117 (2010). 29. Gao, Q., Liu, L., Qian, Q. & Wang, Y. Advances in molecular genetic studies of attention deficit hyperactivity disorder in China. Shanghai Arch Psychiatry. 26, 194–206 (2014). 30. Hawi, Z. et al. The molecular genetic architecture of attention deficit hyperactivity disorder. Mol Psychiatry. 20, 289–297 (2015). 31. Li, Z., Chang, S. H., Zhang, L. Y., Gao, L. & Wang, J. Molecular genetic studies of ADHD and its candidate genes: a review. Psychiatry Res. 219, 10–24 (2014). 32. Euesden, J., Lewis, C. M. & O’Reilly, P. F. PRSice: Polygenic Risk Score software. Bioinformatics. 31, 1466–1468 (2015). 33. Contini, V. et al. Adrenergic α2A receptor gene is not associated with methylphenidate response in adults with ADHD. Eur Arch Psychiatry Clin Neurosci. 261, 205–211 (2011). 34. Neale, B. M. et al. Case-control genome-wide association study of attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc Psychiatry. 49, 906–920 (2010). 35. Lesch, K. P. et al. Molecular genetics of adult ADHD: converging evidence from genome-wide association and extended pedigree linkage studies. J Neural Transm (Vienna). 115, 1573–1585 (2008). 36. Weber, H. et al. Cross-disorder analysis of bipolar risk genes: further evidence of DGKH as a risk gene for bipolar disorder, but also unipolar depression and adult ADHD. Neuropsychopharmacology. 36, 2076–2085 (2011). 37. Barley, K., Dracheva, S. & Byne, W. Subcortical oligodendrocyte- and astrocyte-associated gene expression in subjects with schizophrenia, major depression and bipolar disorder. Schizophr Res. 112, 54–64 (2009). 38. Kurian, S. M. et al. Identification of blood biomarkers for psychosis using convergent functional genomics. Mol Psychiatry. 16, 37–58 (2011). 39. Lyon, G. J. et al. Exome sequencing and unrelated findings in the context of complex disease research: ethical and clinical implications. Discov Med. 12, 41–55 (2011). 40. Penzes, P. & Jones, K. A. Dendritic spine dynamics–a key role for kalirin-7. Trends Neurosci. 31, 419–427 (2008). 41. Park, C., Falls, W., Finger, J. H., Longo-Guess, C. M. & Ackerman, S. L. Deletion in Catna2, encoding alpha N-catenin, causes cerebellar and hippocampal lamination defects and impaired startle modulation. Nat Genet. 31, 279–284 (2002). 42. Conzelmann, A. et al. Methylphenidate normalizes emotional processing in adult patients with attention-deficit/hyperactivity disorder: preliminary findings. Brain Res. 1381, 159–166 (2011). 43. Conzelmann, A. et al. Methylphenidate and emotional-motivational processing in attention-deficit/hyperactivity disorder. J Neural Transm (Vienna). 123, 971–979 (2016). 44. Lasky-Su, J. et al. Genome-wide association scan of quantitative traits for attention deficit hyperactivity disorder identifies novel associations and confirms candidate gene associations. Am J Med Genet B Neuropsychiatr Genet. 147B, 1345–1354 (2008). 45. Zhou, K. et al. Meta-analysis of genome-wide linkage scans of attention deficit hyperactivity disorder. Am J Med Genet B Neuropsychiatr Genet. 147B, 1392–1398 (2008). 46. Arias-Vasquez, A. et al. CDH13 is associated with working memory performance in attention deficit/hyperactivity disorder. Genes Brain Behav. 10, 844–851 (2011). 47. Salatino-Oliveira, A. et al. Cadherin-13 gene is associated with hyperactive/impulsive symptoms in attention/deficit hyperactivity disorder. Am J Med Genet B Neuropsychiatr Genet. 168B, 162–169 (2015). 48. Drgon, T. et al. Genome-wide association for nicotine dependence and smoking cessation success in NIH research volunteers. Mol Med. 15, 21–27 (2009). 49. Treutlein, J. & Rietschel, M. Genome-wide association studies of alcohol dependence and substance use disorders. Curr Psychiatry Rep. 13, 147–155 (2011). 50. Castellanos, F. X. et al. Developmental trajectories of brain volume abnormalities in children and adolescents with attention-deficit/ hyperactivity disorder. JAMA. 288, 1740–1748 (2002). 51. Cortese, S. The neurobiology and genetics of Attention-Deficit/Hyperactivity Disorder (ADHD): what every clinician should know. Eur J Paediatr Neurol. 16, 422–433 (2012). 52. Hoogman, M. et al. Subcortical brain volume differences in participants with attention deficit hyperactivity disorder in children and adults: a cross-sectional mega-analysis. Lancet Psychiatry. 4, 310–319 (2017). 53. Kasparek, T., Theiner, P. & Filova, A. Neurobiology of ADHD From Childhood to Adulthood: Findings of Imaging Methods. J Atten Disord. 19, 931–943 (2015). 54. Rubia, K. et al. Methylphenidate normalises activation and functional connectivity deficits in attention and motivation networks in medication-naive children with ADHD during a rewarded continuous performance task. Neuropharmacology. 57, 640–652 (2009). 55. Rubia, K. et al. Methylphenidate normalizes fronto-striatal underactivation during interference inhibition in medication-naive boys with attention-deficit hyperactivity disorder. Neuropsychopharmacology. 36, 1575–1586 (2011). 56. Rubia, K., Halari, R., Mohammad, A. M., Taylor, E. & Brammer, M. Methylphenidate normalizes frontocingulate underactivation during error processing in attention-deficit/hyperactivity disorder. Biol Psychiatry. 70, 255–262 (2011). 57. He, H. et al. Phosphatidylethanolamine binding protein 4 (PEBP4) is a secreted protein and has multiple functions. Biochim Biophys Acta. 1863, 1682–1689 (2016). 58. Hengst, U., Albrecht, H., Hess, D. & Monard, D. The phosphatidylethanolamine-binding protein is the prototype of a novel family of serine protease inhibitors. J Biol Chem. 276, 535–540 (2001). 59. Hohman, T. J. et al. Discovery of gene-gene interactions across multiple independent data sets of late onset Alzheimer disease from the Alzheimer Disease Genetics Consortium. Neurobiol Aging. 38, 141–150 (2016). 60. George, A. J. et al. A serial analysis of gene expression profile of the Alzheimer’s disease Tg2576 mouse model. Neurotox Res. 17, 360–379 (2010). 61. George, A. J. et al. Decreased phosphatidylethanolamine binding protein expression correlates with Abeta accumulation in the Tg2576 mouse model of Alzheimer’s disease. Neurobiol Aging. 27, 614–623 (2006). 62. Maki, M. et al. Decreased expression of hippocampal cholinergic neurostimulating peptide precursor protein mRNA in the hippocampus in Alzheimer disease. J Neuropathol Exp Neurol. 61, 176–185 (2002). 63. Kobeissy, F. H. et al. Psychoproteomic analysis of rat cortex following acute methamphetamine exposure. J Proteome Res. 7, 1971–1983 (2008). 64. Wei, Q. H. et al. Involvement of hippocampal phosphatidylethanolamine-binding protein in morphine dependence and withdrawal. Addict Biol. 18, 230–240 (2013). SCIENTIFIC REPORTS | (2018) 8:1881 | DOI:10.1038/s41598-018-20194-7 9 www.nature.com/scientificreports/ 65. Lugowska, A. et al. A homozygote for the c.459+1G>A mutation in the ARSA gene presents with cereballar ataxia as the only first clinical sign of metachromatic leukodystrophy. J Neurol Sci. 338, 214–217 (2014). 66. Prados, J. et al. Borderline personality disorder and childhood maltreatment: a genome-wide methylation analysis. Genes Brain Behav. 14, 177–188 (2015). 67. Rybakin, V. et al. Coronin 7, the mammalian POD-1 homologue, localizes to the Golgi apparatus. FEBS Lett. 573, 161–167 (2004). 68. Czibere, L. et al. Profiling trait anxiety: transcriptome analysis reveals cathepsin B (Ctsb) as a novel candidate gene for emotionality in mice. PLoS One. 6, e23604 (2011). 69. Krawitz, P. M. et al. PGAP2 mutations, affecting the GPI-anchor synthesis pathway, cause hyperphosphatasia with mental retardation syndrome. Am J Hum Genet. 92, 584–589 (2013). 70. Makoff, A., Pilling, C., Harrington, K. & Emson, P. Human metabotropic glutamate receptor type 7: molecular cloning and mRNA distribution in the CNS. Brain Res Mol Brain Res. 40, 165–170 (1996). 71. Elia, J. et al. Genome-wide copy number variation study associates metabotropic glutamate receptor gene networks with attention deficit hyperactivity disorder. Nat Genet. 44, 78–84 (2011). 72. Park, S. et al. Association between the GRM7 rs3792452 polymorphism and attention deficit hyperactivity disorder in a Korean sample. Behav Brain Funct. 9, 1 (2013). 73. Yang, L. et al. Polygenic transmission and complex neuro developmental network for attention deficit hyperactivity disorder: genome-wide association study of both common and rare variants. Am J Med Genet B Neuropsychiatr Genet. 162B, 419–430 (2013). 74. Myer, N. M., Boland, J. R. & Faraone, S. V. Pharmacogenetics predictors of methylphenidate efficacy in childhood ADHD. Mol Psychiatry. E-pub ahead of print 12 December. https://doi.org/10.1038/mp.2017.234 (2017). 75. Contini, V. et al. Pharmacogenetics of response to methylphenidate in adult patients with Attention-Deficit/Hyperactivity Disorder (ADHD): a systematic review. Eur Neuropsychopharmacol. 23, 555–560 (2013). Acknowledgements We are grateful to patients from the Hospital Universitari Vall d’Hebron and the Adult ADHD Outpatient Clinic of the Hospital de Clínicas de Porto Alegre, who kindly participated in this research. Genotyping was performed at the Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard, Cambridge, Massachusetts, United States of America. Statistical analyses were carried out on the Genetic Cluster Computer (http://www.geneticcluster.org) hosted by SURFsara and financially supported by the Netherlands Scientific Organization (NWO 480-05-003 PI: Posthuma) along with a supplement from the Dutch Brain Foundation and the VU University Amsterdam. Over the course of this investigation, M.P. has been a recipient of a predoctoral fellowship from the Vall d’Hebron Research Institute (PRED-VHIR-2013) and a research grant from the Deutscher Akademischer Austauschdienst (DAAD), Germany (Research Grants - Short-Term Grants, 2017). C.S.M. is a recipient of a Sara Borrell contract and a mobility grant from the Spanish Ministerio de Economía y Competitividad, Instituto de Salud Carlos III (CD15/00199 and MV16/00039). M.S.A. is a recipient of a contract from the Biomedical Network Research Centre on Mental Health (CIBERSAM), Madrid, Spain. P.R. is a recipient of a pre-doctoral fellowship from the Agència de Gestió d’Ajuts Universitaris i de Recerca (AGAUR), Generalitat de Catalunya, Spain (2016FI_B 00899). I.G.M. is a recipient of a contract from the 7th Framework Programme for Research, Technological Development and Demonstration, European Commission (AGGRESSOTYPE_ FP7HEALTH2013/602805). E.C.S. is a recipient of a pre-doctoral fellowship from the Collaborative Research Training Programme for Medical Doctors (PhD4MD), Institut de Recerca Biomèdica de Barcelona (IRB Barcelona), Spain (II14/00018). M.R. is a recipient of a Miguel de Servet contract from the Instituto de Salud Carlos III, Spain (CP09/00119 and CPII15/00023). This work was funded by Fundación Alicia Koplowitz and Instituto de Salud Carlos III (PI12/01139, PI14/01700, PI15/01789, PI16/01505), and co-financed by the European Regional Development Fund (ERDF), Agència de Gestió d’Ajuts Universitaris i de Recerca-AGAUR, Generalitat de Catalunya, Spain (2014SGR1357, 2014SGR0932), Ministerio de Economía y Competitividad, Spain (SAF2015-68341-R), the European College of Neuropsychopharmacology (ECNP network: ‘ADHD across the lifespan’), Departament de Salut, Generalitat de Catalunya, Spain, and a NARSAD Young Investigator Grant from the Brain & Behavior Research Foundation. The research leading to these results has received funding from the European Union H2020 Programme [H2020/2014-2020] under grant agreements Nos. 667302 (CoCA) and 643051 (MiND). Author Contributions M.P., C.S.M., P.R. and I.G.M. participated in the DNA isolation and preparation of samples. M.P., C.S.M., P.R., M.S.A., I.G.M., B.S.S. and N.R.M. undertook the statistical analyses. V.R., E.C.S., M.C., M.M.V. and E.H.G. contributed to the clinical assessment and recruitment of patients. L.A.R., C.H.D.B., Prof. M.C. and J.A.R.Q. participated in the study design, clinical assessment and coordination of the clinical research. M.R. conceived the project, wrote the protocol and coordinated the study design and the statistical analyses. B.C., J.A.R.Q. and M.R. supervised the project and the manuscript preparation. All authors contributed to and have approved the final version. Additional Information Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-018-20194-7. Competing Interests: E.H.G. has served on the speakers’ bureau and has received travel grants from Shire and Novartis. He has also been on the advisory board and acted as a consultant for Shire. L.A.R. has served on the speakers’ bureau, acted as a consultant and received grant or research support from Eli Lilly and Co., JanssenCilag, Medice, Novartis, and Shire. The ADHD and Juvenile Bipolar Disorder Outpatient Programs chaired by L.A.R. have received unrestricted educational and research support from the following pharmaceutical companies: Eli Lilly and Co., Janssen-Cilag, Novartis, and Shire. L.A.R. has received travel grants from Shire to take part in the 2014 APA, 2015 WFADHD and 2016 AACAP congresses. He has received royalties from Artmed Editora and Oxford University Press. Prof. M.C. has received travel grants and research support from Eli Lilly and Co., Janssen-Cilag, Shire, and Laboratorios Rubió. He has been on the advisory board and served SCIENTIFIC REPORTS | (2018) 8:1881 | DOI:10.1038/s41598-018-20194-7 10 www.nature.com/scientificreports/ as a consultant for Eli Lilly and Co., Janssen-Cilag, Shire, and Laboratorios Rubió. J.A.R.Q. has served on the speakers’ bureau and acted as a consultant for Eli Lilly and Co., Janssen-Cilag, Novartis, Lundbeck, Shire, Ferrer, and Laboratorios Rubió. He has received travel awards from Eli Lilly and Co., Janssen-Cilag, and Shire for participating in psychiatric meetings. The ADHD Program chaired by J.A.R.Q. has received unrestricted educational and research support from Eli Lilly and Co., Janssen-Cilag, Shire, Rovi, and Laboratorios Rubió in the past two years. The remaining authors declare no conflict of interest. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. © The Author(s) 2018 SCIENTIFIC REPORTS | (2018) 8:1881 | DOI:10.1038/s41598-018-20194-7 11